Next-token prediction with the logarithmic loss is a cornerstone of autoregressive sequence modeling, but, in practice, suffers from error amplification, where errors in the model compound and generation quality degrades as sequence length $H$ increases. From a theoretical perspective, this phenomenon should not appear in well-specified settings, and, indeed, a growing body of empirical work hypothesizes that misspecification, where the learner is not sufficiently expressive to represent the target distribution, may be the root cause. Under misspecification -- where the goal is to learn as well as the best-in-class model up to a multiplicative approximation factor $C\geq 1$ -- we confirm that $C$ indeed grows with $H$ for next-token prediction, lending theoretical support to this empirical hypothesis. We then ask whether this mode of error amplification is avoidable algorithmically, computationally, or information-theoretically, and uncover inherent computational-statistical tradeoffs. We show: (1) Information-theoretically, one can avoid error amplification and achieve $C=O(1)$. (2) Next-token prediction can be made robust so as to achieve $C=\tilde O(H)$, representing moderate error amplification, but this is an inherent barrier: any next-token prediction-style objective must suffer $C=\Omega(H)$. (3) For the natural testbed of autoregressive linear models, no computationally efficient algorithm can achieve sub-polynomial approximation factor $C=e^{(\log H)^{1-\Omega(1)}}$; however, at least for binary token spaces, one can smoothly trade compute for statistical power and improve on $C=\Omega(H)$ in sub-exponential time. Our results have consequences in the more general setting of imitation learning, where the widely-used behavior cloning algorithm generalizes next-token prediction.

Holdout validation and hyperparameter tuning from data is a long-standing problem in offline reinforcement learning (RL). A standard framework is to use off-policy evaluation (OPE) methods to evaluate and select the policies, but OPE either incurs exponential variance (e.g., importance sampling) or has hyperparameters on their own (e.g., FQE and model-based). In this work we focus on hyperparameter tuning for OPE itself, which is even more under-investigated. Concretely, we select among candidate value functions ("model-free") or dynamics ("model-based") to best assess the performance of a target policy. Our contributions are two fold. We develop: (1) new model-free and model-based selectors with theoretical guarantees, and (2) a new experimental protocol for empirically evaluating them. Compared to the model-free protocol in prior works, our new protocol allows for more stable generation of candidate value functions, better control of misspecification, and evaluation of model-free and model-based methods alike. We exemplify the protocol on a Gym environment, and find that our new model-free selector, LSTD-Tournament, demonstrates promising empirical performance.

Recent work in language modeling has raised the possibility of self-improvement, where a language models evaluates and refines its own generations to achieve higher performance without external feedback. It is impossible for this self-improvement to create information that is not already in the model, so why should we expect that this will lead to improved capabilities? We offer a new perspective on the capabilities of self-improvement through a lens we refer to as sharpening. Motivated by the observation that language models are often better at verifying response quality than they are at generating correct responses, we formalize self-improvement as using the model itself as a verifier during post-training in order to ``sharpen'' the model to one placing large mass on high-quality sequences, thereby amortizing the expensive inference-time computation of generating good sequences. We begin by introducing a new statistical framework for sharpening in which the learner aims to sharpen a pre-trained base policy via sample access, and establish fundamental limits. Then we analyze two natural families of self-improvement algorithms based on SFT and RLHF. We find that (i) the SFT-based approach is minimax optimal whenever the initial model has sufficient coverage, but (ii) the RLHF-based approach can improve over SFT-based self-improvement by leveraging online exploration, bypassing the need for coverage. Finally, we empirically validate the sharpening mechanism via inference-time and amortization experiments. We view these findings as a starting point toward a foundational understanding that can guide the design and evaluation of self-improvement algorithms.

Language model alignment methods, such as reinforcement learning from human feedback (RLHF), have led to impressive advances in language model capabilities, but existing techniques are limited by a widely observed phenomenon known as overoptimization, where the quality of the language model plateaus or degrades over the course of the alignment process. Overoptimization is often attributed to overfitting to an inaccurate reward model, and while it can be mitigated through online data collection, this is infeasible in many settings. This raises a fundamental question: Do existing offline alignment algorithms make the most of the data they have, or can their sample-efficiency be improved further? We address this question with a new algorithm for offline alignment, $\chi^2$-Preference Optimization ($\chi$PO). $\chi$PO is a one-line change to Direct Preference Optimization (DPO; Rafailov et al., 2023), which only involves modifying the logarithmic link function in the DPO objective. Despite this minimal change, $\chi$PO implicitly implements the principle of pessimism in the face of uncertainty via regularization with the $\chi^2$-divergence -- which quantifies uncertainty more effectively than KL-regularization -- and provably alleviates overoptimization, achieving sample-complexity guarantees based on single-policy concentrability -- the gold standard in offline reinforcement learning. $\chi$PO's simplicity and strong guarantees make it the first practical and general-purpose offline alignment algorithm that is provably robust to overoptimization.

The theory of reinforcement learning (RL) in large state spaces has seen fast development. In the model-free regime, how to use powerful function approximation to learn value functions has been extensively studied in both the online and the offline settings (Jiang et al., 2017; Jin et al., 2020b,c; Xie et al., 2021), which also builds the theoretical foundations that connect RL with (discriminative) supervised learning. On the other hand, generative models for unsupervised/self-supervised learning--which define a sampling distribution explicitly or implicitly--are becoming increasingly powerful (Devlin et al., 2018; Goodfellow et al., 2020), yet how to leverage them to address the key challenges in RL remains under-investigated. While prior works on RL with unsupervised-learning oracles exist (Du et al., 2019; Feng et al., 2020), they often consider models such as block MDPs, which are more restrictive than typical model structures considered in the value-based setting such as low-rank MDPs. In this paper, we study model-free RL in low-rank MDPs with density features for state occupancy estimation. In a low-rank MDP, the transition matrix can be factored into the product of two matrices, and the left matrix is known to serve as powerful features for value-based learning (Jin et al., 2020b), as it can be used to approximate the Bellman backup of any function. On the other hand, the right matrix can be used to represent the policies' state-occupancy distributions, yet how to leverage such density features (without the knowledge of the left matrix) in offline or online RL is unknown.

Standard uniform convergence results bound the generalization gap of the expected loss over a hypothesis class. The emergence of risk-sensitive learning requires generalization guarantees for functionals of the loss distribution beyond the expectation. While prior works specialize in uniform convergence of particular functionals, our work provides uniform convergence for a general class of H\"older risk functionals for which the closeness in the Cumulative Distribution Function (CDF) entails closeness in risk. We establish the first uniform convergence results for estimating the CDF of the loss distribution, yielding guarantees that hold simultaneously both over all H\"older risk functionals and over all hypotheses. Thus licensed to perform empirical risk minimization, we develop practical gradient-based methods for minimizing distortion risks (widely studied subset of H\"older risks that subsumes the spectral risks, including the mean, conditional value at risk, cumulative prospect theory risks, and others) and provide convergence guarantees. In experiments, we demonstrate the efficacy of our learning procedure, both in settings where uniform convergence results hold and in high-dimensional settings with deep networks.

Sample-efficiency guarantees for offline reinforcement learning (RL) often rely on strong assumptions on both the function classes (e.g., Bellman-completeness) and the data coverage (e.g., all-policy concentrability). Despite the recent efforts on relaxing these assumptions, existing works are only able to relax one of the two factors, leaving the strong assumption on the other factor intact. As an important open problem, can we achieve sample-efficient offline RL with weak assumptions on both factors? In this paper we answer the question in the positive. We analyze a simple algorithm based on the primal-dual formulation of MDPs, where the dual variables (discounted occupancy) are modeled using a density-ratio function against offline data. With proper regularization, we show that the algorithm enjoys polynomial sample complexity, under only realizability and single-policy concentrability. We also provide alternative analyses based on different assumptions to shed light on the nature of primal-dual algorithms for offline RL.

To evaluate prospective contextual bandit policies when experimentation is not possible, practitioners often rely on off-policy evaluation, using data collected under a behavioral policy. While off-policy evaluation studies typically focus on the expected return, practitioners often care about other functionals of the reward distribution (e.g., to express aversion to risk). In this paper, we first introduce the class of Lipschitz risk functionals, which subsumes many common functionals, including variance, mean-variance, and conditional value-at-risk (CVaR). For Lipschitz risk functionals, the error in off-policy risk estimation is bounded by the error in off-policy estimation of the cumulative distribution function (CDF) of rewards. Second, we propose Off-Policy Risk Assessment (OPRA), an algorithm that (i) estimates the target policy's CDF of rewards; and (ii) generates a plug-in estimate of the risk. Given a collection of Lipschitz risk functionals, OPRA provides estimates for each with corresponding error bounds that hold simultaneously. We analyze both importance sampling and variance-reduced doubly robust estimators of the CDF. Our primary theoretical contributions are (i) the first concentration inequalities for both types of CDF estimators and (ii) guarantees on our Lipschitz risk functional estimates, which converge at a rate of O(1/\sqrt{n}). For practitioners, OPRA offers a practical solution for providing high-confidence assessments of policies using a collection of relevant metrics.

In order to model risk aversion in reinforcement learning, an emerging line of research adapts familiar algorithms to optimize coherent risk functionals, a class that includes conditional value-at-risk (CVaR). Because optimizing the coherent risk is difficult in Markov decision processes, recent work tends to focus on the Markov coherent risk (MCR), a time-consistent surrogate. While, policy gradient (PG) updates have been derived for this objective, it remains unclear (i) whether PG finds a global optimum for MCR; (ii) how to estimate the gradient in a tractable manner. In this paper, we demonstrate that, in general, MCR objectives (unlike the expected return) are not gradient dominated and that stationary points are not, in general, guaranteed to be globally optimal. Moreover, we present a tight upper bound on the suboptimality of the learned policy, characterizing its dependence on the nonlinearity of the objective and the degree of risk aversion. Addressing (ii), we propose a practical implementation of PG that uses state distribution reweighting to overcome previous limitations. Through experiments, we demonstrate that when the optimality gap is small, PG can learn risk-sensitive policies. However, we find that instances with large suboptimality gaps are abundant and easy to construct, outlining an important challenge for future research.

We cast visual imitation as a visual correspondence problem. Our robotic agent is rewarded when its actions result in better matching of relative spatial configurations for corresponding visual entities detected in its workspace and teacher's demonstration. We build upon recent advances in Computer Vision,such as human finger keypoint detectors, object detectors trained on-the-fly with synthetic augmentations, and point detectors supervised by viewpoint changes and learn multiple visual entity detectors for each demonstration without human annotations or robot interactions. We empirically show the proposed factorized visual representations of entities and their spatial arrangements drive successful imitation of a variety of manipulation skills within minutes, using a single demonstration and without any environment instrumentation. It is robust to background clutter and can effectively generalize across environment variations between demonstrator and imitator, greatly outperforming unstructured non-factorized full-frame CNN encodings of previous works.